Skip to content

Ten Simple Rules for Writing and Sharing Computational Analyses in Jupyter Notebooks

License

Notifications You must be signed in to change notification settings

jupyter-guide/ten-rules-jupyter

Repository files navigation

Ten Simple Rules for Writing and Sharing Computational Analyses in Jupyter Notebooks

Build Status GitHub License

This repository is a supplement to

"Ten Simple Rules for Writing and Sharing Computational Analyses in Jupyter Notebooks"

Tweets

Twitter Tweets

and its preprint

"Ten Simple Rules for Reproducible Research in Jupyter Notebook"

Tweets

Twitter Tweets

Explore the example notebooks below for an application of the Ten Simple Rules.

In addition, we have setup jupyter-guide to crowdsource more technical and in-depth tutorials and to keep up with the rapidly evolving Jupyter ecosystem. We encourage you to contribute and share your expertise.

Example 1

This example demonstrates a reproducible 4-step workflow for predicting a protein fold classification using a Machine Learning approach.


Rule 9: Design Your Notebooks to Be Read, Run, and Explored. The nbviewer links below provide a static preview of notebooks. Click on the launch binder buttons to launch Jupyter Lab in your web browser using the Binder (mybinder.org) server (may be slow!). The HTML links provide a permanent static record of the notebooks. All notebooks can also be launched directly from the links in the 0-Workflow.ipynb top-level notebook.


Nbviewer Jupyter Lab HTML
0-Workflow.ipynb Binder HTML
1-CreateDataset.ipynb Binder HTML
2-CalculateFeatures.ipynb Binder HTML
3-FitModel.ipynb Binder HTML
4-Predict.ipynb Binder HTML

Rule 8: Share and Explain Your Data. To enable reproducibility, we provide a example1/data directory with all data required to run the workflow. A description of the data with download location and download date is available.


Example 2

This example demonstrates a reproducible 2-step workflow for simulating a phylogenetic tree and sequences.

Nbviewer Jupyter Lab HTML
0-Workflow.ipynb Binder HTML
1-SimulateTree.ipynb Binder HTML
2-SimulateSequences.ipynb Binder HTML

Run Jupyter Lab on a local Machine (Laptop, Desktop)

Prerequisites: Miniconda3 (light-weight, preferred) or Anaconda3 and Mamba

  • Install Miniconda3
  • Install Mamba: conda install mamba -n base -c conda-forge

  1. Clone this git repository
git clone https://github.com/jupyter-guide/ten-rules-jupyter.git
  1. Change into the directory
cd ten-rules-jupyter
  1. Create CONDA environment
mamba env create -f environment.yml
  1. Activate the CONDA environment
conda activate ten-rules-jupyter
  1. Launch Jupyter Lab
jupyter lab
  1. Deactivate the CONDA environment
conda deactivate

To remove the CONDA environment, run conda env remove -n ten-rules-jupyter


Citation

Rule A, Birmingham A, Zuniga C, Altintas I, Huang SC, et al. (2019) Ten simple rules for writing and sharing computational analyses in Jupyter Notebooks. PLOS Computational Biology 15(7): e1007007. https://doi.org/10.1371/journal.pcbi.1007007

Contact Us

If you encounter any problems with this repository, please report them here.

About

Ten Simple Rules for Writing and Sharing Computational Analyses in Jupyter Notebooks

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published